The Role of Anger in Predicting Depression and Anxiety Among Individuals with Mental Illness after Psychiatric Hospital Discharge

Background/Purpose The occurrences of anger (or hostility) in individuals experiencing depression and anxiety have been known for many years—since early psychoanalytic theories conceived of anger, or conflict with anger, as one of the basic pathogenetic mechanisms of depression. Anger, as a psychological construct, is conceptualized as an emotional state as well as a personality trait. However, there are inadequate longitudinal studies to differentiate the relative roles of state anger from trait anger as related to depression and anxiety. Therefore, anger is decomposed into state and trait components to test the separate longitudinal effects on depression and anxiety among discharged psychiatric patients.

Methods This research used the MacArthur Violence Risk Assessment Study of 1134 post-discharge psychiatric patients. Depression and anxiety were assessed using the Brief Psychiatric Rating Scale. Anger was measured by 10 items derived from the Novaco Anger Scale at study baseline, 10 weeks, and 30 weeks post-discharge. In a factor analysis conducted by Jones and colleagues (2003), this subscale was labeled as "retaliatory hostility." Internal consistency for this scale ranged from .845 to .857 for each wave in this study. Hierarchical Linear Modeling techniques were employed within this analysis by constructing individual linear growth curve models. To capture the compositional effects of state and trait anger in these models, the time-varying anger was person (or group) mean-centered at Level 1 with reintroduction of the aggregates (of anger) at Level 2 adjusting for marriage, age, SES, gender, and race. This strategy provides unbiased estimates of the orthogonal effects of state (time-varying) and trait (time-invariant) anger on depression and anxiety.

Results Increases in anger significantly predicted changes in depression among discharged patients. Specifically, trait anger was significant for changes in depression over the course of study (B = .01, p <.05). State anger was not significantly related to changes in depression. Both trait and state anger were unrelated to changes in anxiety.

Conclusions and Implications Discharge from psychiatric hospitals is a critical point for patients, as they may be at greater risk of experiencing discontinuity of care. The transition itself may be experienced as a stressful life event as patients need to re-adapt to community environment after living in the structured environment of a hospital. The link between anger and depression may be more accurately explained by considering anger a temperament rather than an emotional state for patients during this particular period. However, it also affects how individuals adapt to the situational demands of stressful life events. It is possible that people with angry and non-adapting temperament style are more prone to run into adverse life events, which, in turn, acts as triggers for depression. Taken together, these findings hold implications for ensuring continuity of care that are particularly compelling when considering potential interventions for anger control (or anger management) and mood disorders among psychiatric patients during the transition to community life.